This is one of three collaborative projects designed to study the genetic epidemiology of breast cancer (BC). The two laboratory-based projects and this project based on family studies and case-control analyses all focus on a population-based cohort of 1400 women who were diagnosed with BC at age less than 45 and a group of women of similar age without BC. These women were interviewed in a case-control study covering diagnosis years 1983-92. DNA has been collected on over half these women and 200 controls. We plan to update and extend the pedigrees of these women and complete blood collections on all cases and an additional 400 controls. Female relatives of 550 cases and 550 controls will be sent mail questionnaires to collect information on the following environmental exposures: reproductive and medical history, hormone use, alcohol use, radiation exposure, body size, diet, as well as other selected exposures that may be related to breast cancer. This information will be used in analyses to separate the environmental and genetic contributions to the familial aggregation of BC and to assess gene/environmental interactions. Should one or more BC genes be identified during the time frame of these studies and if it is possible to test these women for the gene(s), we will be able to search for other BC gene effects by repeating our aggregation and segregation analyses on families which are not affected by the discovered genes. In addition we would be able to directly assess gene-environment interactions using a case-control approach. This cohort of women and their families with the associated information and biologic samples will provide a resource for future genetic studies of BC. This study is one of the first studies that assesses the familial aggregation of BC and segregation of BC genes after controlling for environmental risk factors. Through segregation analysis, we will be able to characterize the underlying gene in terms of its inheritance mode, penetrance and transmission. Similarly, by testing the gene/environmental interactions, we may be able to identify important risk factors in genetically predisposed individuals, thereby providing some guidance for future cancer prevention efforts. This project will interact with the two laboratory projects by working on and sharing results from studies on the same population. Investigators of this project will provide the epidemiologic and biostatistical support to the laboratory studies and will integrate the laboratory results with the environmental determinants.